Human-in-the-loop control strategies in which the user performs a task better, and feels more confident to do so, is an important area of research in teleoperated robotics. However, human behavior can often change as a result of environmental, physical, emotional, and social factors. The goal of this Faculty Early Career Development Program (CAREER) project is to design adaptive control systems that can interpret and react to the dynamic human user. This research will be impactful in the field of surgical robotics where interaction with the patient demands safety and effectiveness from both the human operator (surgeon) and the robotic system itself. Robotic systems that are aware and responsive to user skill and performance style could be more able to avoid user errors and respond to adverse events in unpredictable environments. By integrating real-time models of user intent, movement style, and expertise level with a surgical robotic platform, this project will advance the NSF mission to promote the progress of science and advance national health by exploring fundamental relationships human behavior, motor control, and machine manipulation within the context of surgical robotics. The project supports education and broadening participation in engineering by promoting innovation activities related to healthcare and technology development for medical simulation and training.
The goal of this CAREER project is to develop adaptive control algorithms for teleoperated robotic surgical systems that can respond to, ignore, and/or augment human motor control inputs depending on the output of user-centric models of behavior and task difficulty. Model output will be based in real-time, data-driven predictions and interpretations of human intent, surgical style, and level of expertise. Research objectives include: developing methods to model and control human behavior (e.g., user behavior and expertise) during unstructured teleoperation tasks; designing and analyzing adaptive control laws to enhance performance through visual and haptic guidance; and evaluating the effectiveness of these algorithms on clinically-relevant outcomes in training and intervention using a surgical robotic platform. A key innovation of this work is designing control methods to be agnostic of the specific task performed by the human operator: only user-centric metrics and movement data will serve as inputs to novel difficulty and stylistic prediction models that will then be used to create adaptive control algorithms. This work could lead to significant improvements in the adaptability, capability, and usability of teleoperated surgical systems when collaborating with a human user.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
